A novel way to massively improve the SQ of computer audio streaming

[seeteeyou]

Sound Quality is a Corollary for Measured RF Energy - Part 1

https://audiowise-canada.myshopify.com/blogs/news/measuring-rf-energy

https://audiowise-canada.myshopify.com/collections/cables/products/opto-dx-optical-patch-cable

Quote

Increasing distance reduces measured radiated RF noise since it attenuates according to the square of the distance. Doubling the distance reduces the RF noise by a factor of 4x.

 

@dmance

 

Does it actually mean that we could really use very, VERY LONG optical patch cables until we're hitting some kinda law of diminishing returns? In other words, does it make sense to go beyond 10 meters?

[ray-dude]

Dan can chime in with a lot more pragmatic experiences, but at least in my living room, when using the Opto*DX, there are signal integrity issues as you extend the length of the optical cables.

 

Background on Opto•DX in my long review linked below, but it is essentially a passive transceiver for coax-SPDIF waveforms to optical back to coax-SPDIF waveforms (no reclocking or signal regeneration).  The Chord DAVE is more sensitive to SPDIF waveform for good dual SPDIF (DX) sync than other Chord DACs.  When Dan was visiting San Diego several months ago, we were limited to 3m for having stable DX sync between Chord Hugo mScaler and Chord DAVE.  I only recently got a Chord TT2, but Dan has shared that with his TT2 he can go a lot longer with the optical runs and stil have a stable DX sync (5-10m).  However, even then, the wave form does devolve, and you do run into sync issues.

 

An open question for me is how the DPLL in Chord DACs respond to signal integrity degradation in the DX signal.  Rob is obviously reclocking the signal internal to the DACs, but where is the boundary where the benefits of RF reduction (additional physical distance) are outweighed by SQ degradation due to stress on the DPLL?

 

I do have a 10m optical patch cable in house (along with my 3m cable). so I hope to get some practical listening tests in over the next several weeks (and I'm also looking forward to hearing Dan's RF•Block product: https://audiowise-canada.myshopify.com/products/rf-block )  

 

 

Opto•DX review and background:

 

https://www.head-fi.org/showcase/audiowise-opto•dx-optical-isolation-bridge-for-dual-spdif.23757/reviews#review-22155

 

[soares]

Hi folks!

Does anyone knows if I can use the Apacer memory on an Innous Zen mkIII?

 Thank you,

Jorge

[seeteeyou]

https://dougbradyhifi.com/products/innuos-zen-mk-ii

Quote

8GB DDR3 Low-Voltage RAM (4GB Dedicated for RAM Playback)

 

Take a look at these links below, Innuos seemed to have something very similar to Supermicro X11SSA with a few things removed

 

https://www.hifitest.de/test/bildergalerie/musikserver/innuos-zen_mk_iii_17558/5

https://www.hifitest.de/test/bildergalerie/musikserver/innuos-zen_mk_iii_17558/7

https://audiophilestyle.com/forums/topic/55834-innuos-zenith-mk3-teardown/?tab=comments#comment-941020

 

That means something like this with Wide Temp. might work

 

http://eflash.apacerus.com/spec/DRAM/I-Grade/75.C93E2.G000C.pdf

https://industrial.apacer.com/en-ww/DRAM/DDR3-Wide-Temp.-SODIMM

 

8GB for €58

 

https://www.rutronik24.com/product/apacer/75.c93e2.g040b/8434061.html

https://www.mouser.com/ProductDetail/Apacer/75C93E2G000C?qs=P%2FC4Eoq9CDxJOsd716Bv0g%3D%3D

[seeteeyou]

FYI - current owners of Mutec REF10 might wanna keep an eye on upgrades

 

https://www.hiendy.com/hififorum/forum.php?mod=viewthread&tid=124319&page=49#pid2976776

Quote

Mutec Ref10 SE-120 absolutely takes the already exceptional dCS Vivaldi Clock to a whole new level !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

 

https://www.whatsbestforum.com/threads/reality-is-cruel-cybershaft-new-ultimate-ocxo-10m-clocks-shootout-op20-vs-op17.26610/page-8#post-612931

Quote

Knowing that the upgrade process may not be available readily, since it is logistically difficult and time consuming and costly to perform such upgrades

 

BTW, brand new units are available for 42,500 RMB

 

https://item.taobao.com/item.htm?id=569160752500

 

 

[seeteeyou]

On 9/23/2019 at 1:02 PM, austinpop said:

There is nothing magical about the Pilot Pro 2. It's still just a battery, and as such, has the usual battery issues of - presumably, as I have not measured it - poor regulation, high output impedance, etc.

 

FYI

 

https://www.ldovr.com/product-p/dxp-1a5s.htm

https://www.head-fi.org/threads/hugo-m-scaler-by-chord-electronics-the-official-thread.885042/page-647#post-15347289

Quote

In my setup, DXP-1A5S powered by the PP2 blows any other power supply that I tried out of the water. To my ears, it sounds clearly better than SR4 and Keces P8 with 15V, Gissaral power bank with 15V and PP2 with 12V.

 

[seeteeyou]

ICYMI - @romaz visited Hengelo and met the owner of Taiko Audio recently

 

https://www.whatsbestforum.com/threads/taiko-audio-sgm-extreme-the-crème-de-la-crème.27433/page-88#post-613373

 

Emile actually designed his own motherboards for SGM Extreme but in the end Asus WS C621 SAGE still came out on top.

[seeteeyou]

On 9/17/2019 at 2:08 AM, Nenon said:

To me, getting I2s from the PCIe slot on the computer is the best way to extract digital audio from the source to my DAC. Given that most DACs don't reclock the I2s signal, it is important to have a good clock at the source. This is where the PinkFaun Ultra OCXO makes a good audible difference. Once you extract the digital bits from your computer, the main thing we need to worry about it is preserving that clock signal over cables to your DAC.

 

Not trying to "spam" this thread with so many replies in a row but this particular one below looked quite intriguing to me

 

https://www.monoandstereo.com/2019/12/pink-faun-streamer-216x-review.html

Quote

The PSU is low-noise, the clocks also, this results in fewer mistakes made by the processor because there's less noise. So the PSU and ULTRA clocks make sure the latency is even lower than with the standard clocks on the motherboard. It's even measurable when the streamer is equipped with ULTRA OCXOs or not. The load on the processor is around 0.2% with the ULTRA clock, with the standard motherboard clock the load on the processor is around 2%

 

Let's wait and see how this guy would perform in the future

 

https://www.fidelityaudio.co.uk/c5-signature-ultra-low-jitter-clock-4461-p.asp

[Confused]

9 hours ago, seeteeyou said:

FYI - current owners of Mutec REF10 might wanna keep an eye on upgrades

 

https://www.hiendy.com/hififorum/forum.php?mod=viewthread&tid=124319&page=49#pid2976776

 

https://www.whatsbestforum.com/threads/reality-is-cruel-cybershaft-new-ultimate-ocxo-10m-clocks-shootout-op20-vs-op17.26610/page-8#post-612931

 

BTW, brand new units are available for 42,500 RMB

 

https://item.taobao.com/item.htm?id=569160752500

 

 

Interesting, although for me I would much rather see an improved version of the MC3+USB (without SMPS) than a slightly lower phase noise version of a very low phase noise clock.  Depends on how you are using it I guess.

[Ricardo007]

1 hour ago, seeteeyou said:

Let's wait and see how this guy would perform in the future

 

https://www.fidelityaudio.co.uk/c5-signature-ultra-low-jitter-clock-4461-p.asp

I plan to have one of their clock modules installed in my ubiquity edge router 6P next year...

Can somebodty share experience on their clocks?

[soares]

10 hours ago, seeteeyou said:

https://dougbradyhifi.com/products/innuos-zen-mk-ii

 

Take a look at these links below, Innuos seemed to have something very similar to Supermicro X11SSA with a few things removed

 

https://www.hifitest.de/test/bildergalerie/musikserver/innuos-zen_mk_iii_17558/5

https://www.hifitest.de/test/bildergalerie/musikserver/innuos-zen_mk_iii_17558/7

https://audiophilestyle.com/forums/topic/55834-innuos-zenith-mk3-teardown/?tab=comments#comment-941020

 

That means something like this with Wide Temp. might work

 

http://eflash.apacerus.com/spec/DRAM/I-Grade/75.C93E2.G000C.pdf

https://industrial.apacer.com/en-ww/DRAM/DDR3-Wide-Temp.-SODIMM

 

8GB for €58

 

https://www.rutronik24.com/product/apacer/75.c93e2.g040b/8434061.html

https://www.mouser.com/ProductDetail/Apacer/75C93E2G000C?qs=P%2FC4Eoq9CDxJOsd716Bv0g%3D%3D

Thank you so much. I’ll give it a try! 🙂

Cheers Jorge

[seeteeyou]

My pleasure, most likely the same Apacer RAM should also work for ZENith Mk.II (SE) with Celeron J3455 as well as Statement with Celeron N4200. Pretty much all of us should have tried DDR4 ones before and that's why it would be great to find out if older DDR3 ones were just as good or otherwise.

 

Please don't forget to let us know how it's working out because other owners of Innuos music servers will most likely join the party, they won't be left out even though DDR4 SODIMM ain't supported by their motherboards.

[justubes]

On 12/8/2019 at 6:56 AM, seeteeyou said:

Does anyone understand if the Ram clocking is obtained from or shared with the CPUs clock?

lockquote widget

[soares]

On 12/8/2019 at 11:58 AM, seeteeyou said:

My pleasure, most likely the same Apacer RAM should also work for ZENith Mk.II (SE) with Celeron J3455 as well as Statement with Celeron N4200. Pretty much all of us should have tried DDR4 ones before and that's why it would be great to find out if older DDR3 ones were just as good or otherwise.

 

Please don't forget to let us know how it's working out because other owners of Innuos music servers will most likely join the party, they won't be left out even though DDR4 SODIMM ain't supported by their motherboards.

Well, know that I saw the prices (250$) I will probably not try it after all, as to start I am not sure what will be the result... 😞

The link you provide with a lower price (https://www.rutronik24.com/product/apacer/75.c93e2.g040b/8434061.html) only sell to companies.

Is there a way to circumvent this limitation?

Thank you,

Jorge 

[austinpop]

DAC Update

 

As I posted a few weeks ago, I needed to experiment where in my system the 3 DR rails of my new Paul Hynes SR-7 PSU had the biggest impact. I'll report on that in my next post. But first, one big change in my system is that I have replaced the Chord TT2 with a DAVE as my DAC. As I reported a couple months ago, when I heard the HMS/DAVE in my system for the first time, the improvement going from even the HMS/TT2 to HMS/DAVE is nothing short of astonishing. 

 

Several people I've corresponded with privately have reminded me of a post I made in mid 2017 about not considering the DAVE to be that big of an improvement then. Here's what I posted then: https://audiophilestyle.com/forums/topic/30376-a-novel-way-to-massively-improve-the-sq-of-computer-audio-streaming/?do=findComment&comment=660539 

 

Result: The DAVE is clearly better. I found this to be more evident than Eric did.  We experimented with some settings on the DAVE, settling on DSD+ and HF Filter ON. I felt the DAVE just presented another layer of detail and clarity than the Codex. Eric felt the impact of the switch in the trifecta was at least as significant as the DAVE. We both agreed the effect of the SOtM stack was more dramatic than the the DAC difference between the Codex and the DAVE, which is notable in itself! Both of us couldn't justify the 6.5x difference in cost, but there is no denying the quality of the DAVE. Oh - to be rich!!

 

I stand by my comments - then and now. With the then-state of my system, the improvement of upstream optimizations like the SOtM trifecta had a bigger impact than DAC improvements. Today, 2.5 years of upstream optimizations later, my system is far better able to showcase DAC improvements. 

 

[marce]

On 12/8/2019 at 2:07 PM, seeteeyou said:

Let's wait and see how this guy would perform in the future

 

https://www.fidelityaudio.co.uk/c5-signature-ultra-low-jitter-clock-4461-p.asp

The PCB layout shown is horrible, it looks like its been autorouted, something that is not done these days... The track coming out of C20 is a prime example, it also looks like a basic 2 layer board... Not the best for a critical design such as a clock...

[austinpop]

PSU Update

 

As I posted a few weeks ago, I needed to experiment where in my system the 3 DR rails of my new Paul Hynes SR-7 PSU had the biggest impact. I've now completed a full round of comparisons and settled on locations. To illustrate the experiment, here is a picture of my latest system topology:

 

 

The experiment was to decide where to put 2 of my SR-7 DR rails, with the 3rd 15V/6A DR rail powering my HMS. The choices were the blue and red circles in the picture.

1. Powering my server (red combo)
   • SR-7 19V/10A DR rail > HDPlex DC-ATX adapter > ATX on the mobo
   • SR-7 12V/6A DR rail > EPS (CPU) on mobo
   • JCAT Net card now bus powered
   • Switch and tX-USBultra powered by 12V/3A SR7 SR rails
      
2. Powering the switch and tX (blue combo)
   • HDPlex 400W ATX LPS powering the server
     • ATX output > ATX on mobo
     • separate 12V/7A rail (modded from the stock 19V XLR output) > EPS on mobo
   • SR-4 5V/2A > JCAT Net Card Femto
   • SR-7 12V/6a DR rail > sNH-10G switch
   • SR-7 12V/6a DR rail > tX-USBultra

Result: While this was not a trivial swap, and took some time to switch from the red to the blue combo and back, the end result was that I found the blue combo, i.e. DR rails on the switch and tX-USBultra, to sound better. So for now, I am keeping my HDPlex 400W Linear ATX supply.

 

Corollary: I was able to borrow @limniscate's SR7MR2DRXL for a few days, and constructed an all-SR7DR super combo chain:

• SR-7 19V/10A DR rail > HDPlex DC-ATX adapter > ATX on the mobo
• SR-7 12V/6A DR rail > EPS (CPU) on mobo
• JCAT Net card now bus powered
• SR-7 12V/6a DR rail > sNH-10G switch
• SR-7 12V/6a DR rail > tX-USBultra

This super-combo just smokes both the red and blue combos! I have a hard time even articulating how good this sounds. The HDPlex Linear ATX supply is already very good, but using the SR-7 to power the server just takes it to another level. SQ improves across the board, in dynamics, transparency, bass articulation, tonality and imaging. I finally got to hear what @romaz has been raving about, and it is truly something! I know the current buzz is about the impossibly expensive (for me) SGM Extreme, but I really have to wonder just how much better the Extreme or Innuos Statement can be over my custom server powered by SR-7 DR rails. Maybe in the not too distant future, we'll be able to find out.

 

I'm now on the hunt for another SR-7 MR2DRXL, so I'm waiting to see if a slot on Paul's custom build queue opens up.

 

Finally, here are some miscellaneous findings from these PSU experiments:

1. HDPlex's bigger, badder 800W DC-ATX adapter is a real SQ champ. It sounds significantly better than the 400W DC-ATX adapter. Thanks again to Roy for this tip. While it does take up more space, and really needs a bigger case than the HDPlex H3, it makes up for it in SQ. The other potential concern is inrush current, if you're trying to use it with a lower current rail. In my case, my 19V/10A rail didn't even break a sweat.
2. PSU SQ is related to the current demand and capacity. By this I mean two things:
   • Just because a PSU "works" under a certain load does not mean it will sound its best
   • PSUs that have lower current capacity tend to underperform as the load approaches their capacity.
   • Several cases in point:
     • When powering the Mytek Brooklyn DAC+ (a current-hungry application), I found the JS-2 to outperform the SR-4, even though in a lower-current application powering the tX-USBultra, the SR-4 outperformed the JS-2. Yes, the SR-4 "worked" in this application, but its nominal 2A sustained max current made it sound more constricted and polite than the JS-2
     • Similarly, when powering the Chord TT-2 with the Farad Super 3 and the SR4-15, the Farad's superior 3A current rating allowed it to sound better than the SR-4. Again, in a lower-current application powering the tX-USBultra, the SR-4 and Farad were neck and neck.
3. Finding 2 is relevant to explain the results of another experiment I tried. Since I also have a 3-rail SR-7 SR, with 15v/5a, 15v/3a, and 15v/2a rails, I wondered if I could use an SR 12v/4a rail (derated current due to voltage reduction) to power my server's CPU via the EPS port. As it turned out, this did not outperform the EPS powered from the 12V/7A rail from the HDPlex. While the SR7 SR rail had some advantages in the form of tonality, refinement, and transparency, it lacked the dynamics of the HDPlex by quite a margin, which tilted the balance in the HDPlex's favor.
4. When powering a music server, muscular current capacity matters in the PSU.

[RickyV]

1 hour ago, austinpop said:

PSU Update

 

As I posted a few weeks ago, I needed to experiment where in my system the 3 DR rails of my new Paul Hynes SR-7 PSU had the biggest impact. I've now completed a full round of comparisons and settled on locations. To illustrate the experiment, here is a picture of my latest system topology:

 

 

The experiment was to decide where to put 2 of my SR-7 DR rails, with the 3rd 15V/6A DR rail powering my HMS. The choices were the blue and red circles in the picture.

1. Powering my server (red combo)
   • SR-7 19V/10A DR rail > HDPlex DC-ATX adapter > ATX on the mobo
   • SR-7 12V/6A DR rail > EPS (CPU) on mobo
   • JCAT Net card now bus powered
   • Switch and tX-USBultra powered by 12V/3A SR7 SR rails
      
2. Powering the switch and tX (blue combo)
   • HDPlex 400W ATX LPS powering the server
     • ATX output > ATX on mobo
     • separate 12V/7A rail (modded from the stock 19V XLR output) > EPS on mobo
   • SR-4 5V/2A > JCAT Net Card Femto
   • SR-7 12V/6a DR rail > sNH-10G switch
   • SR-7 12V/6a DR rail > tX-USBultra

Result: While this was not a trivial swap, and took some time to switch from the red to the blue combo and back, the end result was that I found the blue combo, i.e. DR rails on the switch and tX-USBultra, to sound better. So for now, I am keeping my HDPlex 400W Linear ATX supply.

 

Corollary: I was able to borrow @limniscate's SR7MR2DRXL for a few days, and constructed an all-SR7DR super combo chain:

• SR-7 19V/10A DR rail > HDPlex DC-ATX adapter > ATX on the mobo
• SR-7 12V/6A DR rail > EPS (CPU) on mobo
• JCAT Net card now bus powered
• SR-7 12V/6a DR rail > sNH-10G switch
• SR-7 12V/6a DR rail > tX-USBultra

This super-combo just smokes both the red and blue combos! I have a hard time even articulating how good this sounds. The HDPlex Linear ATX supply is already very good, but using the SR-7 to power the server just takes it to another level. SQ improves across the board, in dynamics, transparency, bass articulation, tonality and imaging. I finally got to hear what @romaz has been raving about, and it is truly something! I know the current buzz is about the impossibly expensive (for me) SGM Extreme, but I really have to wonder just how much better the Extreme or Innuos Statement can be over my custom server powered by SR-7 DR rails. Maybe in the not too distant future, we'll be able to find out.

 

I'm now on the hunt for another SR-7 MR2DRXL, so I'm waiting to see if a slot on Paul's custom build queue opens up.

 

Finally, here are some miscellaneous findings from these PSU experiments:

1. HDPlex's bigger, badder 800W DC-ATX adapter is a real SQ champ. It sounds significantly better than the 400W DC-ATX adapter. Thanks again to Roy for this tip. While it does take up more space, and really needs a bigger case than the HDPlex H3, it makes up for it in SQ. The other potential concern is inrush current, if you're trying to use it with a lower current rail. In my case, my 19V/10A rail didn't even break a sweat.
2. PSU SQ is related to the current demand and capacity. By this I mean two things:
   • Just because a PSU "works" under a certain load does not mean it will sound its best
   • PSUs that have lower current capacity tend to underperform as the load approaches their capacity.
   • Several cases in point:
     • When powering the Mytek Brooklyn DAC+ (a current-hungry application), I found the JS-2 to outperform the SR-4, even though in a lower-current application powering the tX-USBultra, the SR-4 outperformed the JS-2. Yes, the SR-4 "worked" in this application, but its nominal 2A sustained max current made it sound more constricted and polite than the JS-2
     • Similarly, when powering the Chord TT-2 with the Farad Super 3 and the SR4-15, the Farad's superior 3A current rating allowed it to sound better than the SR-4. Again, in a lower-current application powering the tX-USBultra, the SR-4 and Farad were neck and neck.
3. Finding 2 is relevant to explain the results of another experiment I tried. Since I also have a 3-rail SR-7 SR, with 15v/5a, 15v/3a, and 15v/2a rails, I wondered if I could use an SR 12v/4a rail (derated current due to voltage reduction) to power my server's CPU via the EPS port. As it turned out, this did not outperform the EPS powered from the 12V/7A rail from the HDPlex. While the SR7 SR rail had some advantages in the form of tonality, refinement, and transparency, it lacked the dynamics of the HDPlex by quite a margin, which tilted the balance in the HDPlex's favor.
4. When powering a music server, muscular current capacity matters in the PSU.

 

Wow Rajiv very useful post, as always. I have to read it again when I get home.

I was wondering did you consideredthe dCS Bartok instead of the HMS+Dave combo?

 

[bodiebill]

2 hours ago, austinpop said:

1. HDPlex's bigger, badder 800W DC-ATX adapter is a real SQ champ. It sounds significantly better than the 400W DC-ATX adapter. Thanks again to Roy for this tip. While it does take up more space, and really needs a bigger case than the HDPlex H3, it makes up for it in SQ. The other potential concern is inrush current, if you're trying to use it with a lower current rail. In my case, my 19V/10A rail didn't even break a sweat.

 

Could someone explain what are the options to power the HDPlex 800W DC-ATX adapter?

s it possible to use a conventional (nonlinear) PC power supply to do this? Such as my Seasonic 400W fanless platinum PSU?

[rickca]

2 hours ago, austinpop said:

a lower-current application powering the tX-USBultra

What applications are you comparing?  

[Patatorz]

16 minutes ago, bodiebill said:

 

Could someone explain what are the options to power the HDPlex 800W DC-ATX adapter?

s it possible to use a conventional (nonlinear) PC power supply to do this? Such as my Seasonic 400W fanless platinum PSU?

not able to find the 800w dc-Atx on hdplex website...

[rickca]

4 minutes ago, Patatorz said:

not able to find the 800w dc-Atx on hdplex website...

https://hdplex.com/hdplex-800w-dc-atx-with-16v-63vdc-input.html

[rickca]

3 hours ago, austinpop said:

HDPlex's bigger, badder 800W DC-ATX adapter is a real SQ champ.

How did you power this?  Doesn't it require two 6-pin 19V inputs?

[rickca]

1 hour ago, bodiebill said:

 

Could someone explain what are the options to power the HDPlex 800W DC-ATX adapter?

s it possible to use a conventional (nonlinear) PC power supply to do this? Such as my Seasonic 400W fanless platinum PSU?

Conventional ATX power supplies don't have a 19V output required to power the DC-ATX.

[rickca]

3 hours ago, austinpop said:

I found the blue combo, i.e. DR rails on the switch and tX-USBultra, to sound better.

Does this mean in your system the power quality/capacity provided for the switch and tX-USBultra is more critical than that provided for the server, or can you not draw that conclusion from your A vs B comparison?